The nonlinear dynamics of space weather

Abstract

Studies of the nonlinear magnetospheric dynamics have led to several directions useful in understanding space physics processes, in particular those related to magnetospheric currents, and making space weather forecasts possible. Four such directions are identified: (a) empirical time series prediction with nonlinear autoregressive moving-average (ARMA) models, for which an example is given in terms of a geosynchronous electron flux index model. (b) Measurement of physical properties of the currents from the coefficients of the ARMA models, such as characteristic time scales and coupling strengths. Using a D st index model we measure the ring current decay time as a function of storm phase and activity, and identify new oscillation time scales correlating with the substorm injection activity. (c) Spatiotemporal nonlinear modeling predicts the amplitude and location of the disturbance as a function of space, as well as its time evolution. An example is given in terms of predicting the longitudinal and temporal profile of midlatitude geomagnetic disturbances. (d) Nonlinear dynamical models can be coupled to other physical or empirical approaches to build comprehensive space weather models. SWIFT, and ionospheric space weather model, is discussed as an example. Other applications of nonlinear dynamics are briefly discussed.